The Spanish Journal of Psychology
2008, Vol. 11, No. 1, 55-65
Copyright 2008 by The Spanish Journal of Psychology
ISSN 1138-7416
Women’s Access to Managerial Positions: An Experimental
Study of Leadership Styles and Gender
Isabel Cuadrado, J. Francisco Morales, and Patricia Recio
Universidad Nacional de Educación a Distancia
This study attempts to test one of the explanations of the scarce representation of women
in managerial positions, specifically the one advanced by “role congruity theory of prejudice
toward female leaders” (Eagly & Karau, 2002), which appeals to the fact that women get
unfavorable evaluations if they adopt male-stereotypical leadership styles. One-hundred
and thirty-six undergraduate students participated in an experiment with a 2 (Malestereotypical vs. Female-stereotypical leadership style) × 2 (Male vs. Female leader)
design. Dependent variables were leader’s competence, efficacy, and evaluation in a series
of traits. It was found that, regardless of sex, the leaders were considered more competent
and efficient, and were evaluated more favorably, when they adopted stereotypically
feminine leadership styles. Implications of these findings for women’s underrepresentation
as leaders in management top positions worldwide are discussed.
Keywords: glass ceiling, leadership styles, gender, female leaders
El objetivo de este estudio es investigar si, como postula la teoría del prejuicio de rol
hacia líderes femeninos (Eagly y Karau, 2002), una de las causas de la escasa presencia
femenina en puestos directivos es la devaluación de las mujeres líderes cuando adoptan
formas de liderazgo estereotípicamente masculinas. Para ello, diseñamos un experimento
en el que participaron 136 estudiantes. Se elaboró una historia en la que se manipulaba
el estilo de liderazgo (estereotípicamente masculino vs. estereotípicamente femenino) y
el sexo del líder (hombre vs. mujer). Las variables dependientes fueron la capacidad de
liderazgo, la eficacia en el desempeño del liderazgo y la evaluación positiva o negativa
del líder (medida a través de una lista de adjetivos). Los resultados indican que,
independientemente del sexo, los líderes son mejor evaluados en todas las variables
cuando ejercen el liderazgo con estilos estereotípicamente femeninos. Se discuten estos
resultados a partir de la literatura psicosocial relacionada con el acceso limitado de las
mujeres a puestos directivos.
Palabras clave: techo de cristal, estilos de liderazgo, género, mujeres líderes
The authors wish to thank the three anonymous reviewers for their comments on prior versions of this article.
Correspondence concerning this article should be addressed to Isabel Cuadrado Guirado. Dpto. de Psicología Social y de las
Organizaciones. Facultad de Psicología. UNED. C/ Juan del Rosal, 10. 28040. Madrid. E-mail: icuadrado@psi.uned.es
Translation: Virginia Navascués Howard.
55
56
CUADRADO, MORALES, AND RECIO
The relation between leadership styles and gender has
recently become an important topic of research (i.e., Barberá
& Ramos, 2004; Cuadrado, 2003; Cuadrado & Molero,
2002; Eagly & Johnson, 1990; Eagly & Johannesen-Schmidt,
2001; Eagly, Johannesen-Schmidt, & van Engen, 2003; van
Engen & Willemsen, 2004), generating interesting debates
in the literature, both in Spain (see Ayestarán, 2003;
Cuadrado, 2003; Moya, 2003; Munduate, 2003), and beyond
our frontiers (see Eagly & Carli, 2003a, 2003b; Vecchio,
2002, 2003). As noted by Eagly and Carli (2003b), these
studies are sometimes carried out to investigate whether the
scarcity of women in managerial positions can be explained
on the basis of the fact that they use less effective leadership
styles than men, and sometimes, to investigate whether
women use “superior” leadership styles than men’s styles,
which has recently been called the female leadership
advantage (see Eagly & Carli, 2003a, 2003b; Helgesen,
1990; Vecchio, 2002, 2003).
As seen in the pertinent literature (see, for example,
Cuadrado, 2001; Eagly & Johnson, 1990), the main styles
on which classic research has focused are autocratic versus
democratic and task-oriented/relationship-oriented. Autocratic
leadership is characterized by the leader’s making decisions
unilaterally, not allowing the group members to participate.
Democratic leadership is participative, consultative, and
involves the group, and the leader allows and encourages
group members’ participation in the decisions (Cuadrado).
Leaders who use a task-oriented style are mainly concerned
with achieving the group goals—emphasis on achieving the
task—whereas relationship-oriented leaders are basically
concerned with their followers’ wellbeing and satisfaction—
emphasizing the quality of relations with others—(Cuadrado).
As indicated by Eagly and Johnson when referring to these
styles (p. 236), “leadership research provides an excellent
opportunity to determine whether the behavior of leaders is
gender stereotypic.” Men are generally considered more
autocratic and task-oriented because of their relationship
with the components of the “instrumental” dimension of
gender stereotypes (e.g., aggressive, enterprising,
independent, self-sufficient, dominant, competent, rational).
In contrast, women tend to be considered more democratic
and relationship-oriented, because the “communal” dimension
is characterized by aspects such as being concerned with
others, being generous, sensitive, understanding, affectionate,
or compassionate (see Cuadrado, Navas, & Molero, 2006).
Studies carried out by Cann and Siegfried (1990) offer
empirical support to this relation between gender stereotypes
and leadership styles.
In the last decades, attention has focused on gender
differences in “transactional” and “transformational”
leadership (e.g., Cuadrado & Molero, 2002; Eagly et al.,
2003). Transactional leadership is characterized by the
establishment of a kind of “exchange” between the leader
and the group members. Transformational leaders do not
limit themselves to establishing a mere exchange of rewards
with their followers or subordinates, but instead they produce
changes in their followers’ scale of values, attitudes, and
beliefs, thus getting them to attain much higher levels of
achievement than expected and, therefore, higher levels of
job satisfaction (Bass & Avolio, 1990). As noted by Eagly
and Johannesen-Schmidt (2001), although these styles are
not so obviously related to gender as the former ones,
transformational leadership has “communal” aspects,
especially the factor “individualized consideration,” which
make it a style more aligned with women. Individualized
consideration refers to the leader’s capacity to pay personal
attention to all the team members, making them feel that
their individual contribution is important (Bass & Avolio,
1990). Likewise, according to van Engen, van der Leeden,
and Willemsen (2001), transformational leadership can be
considered a “female” style, because of the emphasis on
individualized consideration to subordinates (which is related
to characteristics stereotypically attributed to women). These
authors add that “many authors explicitly refer to this style
as a female leadership style” (p. 583).
Therefore, in view of the above, the autocratic and taskoriented leadership styles are stereotypically male, whereas
the democratic and relationship-oriented styles, with
individualized consideration of team members are considered
stereotypically female (see Cuadrado, 2003; Cuadrado et
al., 2006).
The results from two meta-analyses carried out by Eagly
and collaborators on gender and evaluation of leaders (Eagly,
Makhijani, & Klonsky, 1992), and on gender and
effectiveness of leaders (Eagly, Karau, & Makhijani, 1995)
have important implications for the consequences of each
sex of adopting male or female leadership styles. Thus, the
first of these meta-analyses showed that female leaders
obtained poorer evaluations than males when they adopted
a male-stereotypical leadership style and when the evaluators
were men. On the contrary, male leaders were not evaluated
less favorably than females when they adopted a femalestereotypical leadership style, and female evaluators—in
contrast to the males—did not favor female leaders over
male leaders. Moreover, in the meta-analysis of gender and
effectiveness of leaders, the results showed than male leaders
were more effective than females in stereotypical male
activities (those that require “task” skills: directing and
controlling others), whereas female leaders were more
effective than male leaders in stereotypical female activities
(those that require “interpersonal” skills: cooperating and
getting on well with others). Likewise, when there were a
higher number of men among the evaluators, the male
leaders obtained better evaluations of effectiveness than the
females. However, female evaluators did not favor one sex
over the other in their evaluations of effectiveness.
To sum up, the above results show that female leaders
are at a disadvantage with regard to males when they adopt
male-stereotypical styles and when the evaluators (their
subordinates) are men. Therefore, differences in leadership
WOMEN’S ACCESS TO MANAGERIAL POSITIONS
between men and women are not so much due to the fact
that they act differently but to differential reactions to the
behavior of both sexes. Taking these and other results as
a starting point, Eagly and Karau (2002) formulated the role
congruity theory of prejudice toward female leaders (see
Morales & Cuadrado, 2004, for an analysis in Spanish of
the theory). The central theme of the theory is that the
perceived incongruity between the female stereotype and
leadership roles is responsible for the existence of two kinds
of prejudice towards female leaders and potential leaders,
which results in a poorer appraisal of women’s leadership
behavior in comparison to that of men. The first kind of
prejudice, proceeding from the descriptive component of
the gender stereotypes—beliefs about men’s and women’s
characteristics—is due to the mismatch between leadership
roles and typically female characteristics. The second, derived
from the prescriptive gender stereotypes—beliefs about the
desirable characteristics of the members of each sex—is due
to the fact that stereotypically male behavior is perceived
as less desirable in women than in men. Thus, according to
Eagly and Karau, female leaders or potential leaders receive
fewer favorable evaluations than their males counterparts
when they adopt male-stereotypical leadership styles—for
example, autocratic and task-oriented styles—(see also Eagly
et al., 2003, pp. 571-573).
It is noteworthy that men’s and women’s leadership
behaviors are represented in the literature in a fairly isolated
manner to facilitate their exposition, but in practice, there
is more overlap. Thus, for example, a leader can, at the same
time, be democratic, relationship-oriented, and task-oriented
(see Cuadrado, 2001). However, the studies carried out to
investigate the effects of leadership styles as a function of
gender have not taken this aspect into account, but instead
they have been limited to studying leadership differences
between men and women separately (see Cuadrado, 2003).
In fact, the meta-analysis of Eagly and Johnson (1990) on
gender differences in leadership styles (162 studies that
produce 370 comparisons) showed the extension of the
works on this issue till the present; however, the studies
included focus on the gender differences only in one kind
of style: relationship-oriented/task-oriented, relationshiporiented and task-oriented, and autocratic versus democratic.
Likewise, gender differences have not been investigated
equally in all the styles or with the same methodology, so
that autocratic versus democratic style is the least studied
(n = 28 from 370 comparisons), and the experimental studies
are the least numerous (n = 81 out of a total of 370
comparisons). In general, women use the democratic style
to a greater extent than men, and men use the autocratic
style more frequently than women (see Cuadrado & Navas,
2000, for a more detailed analysis in Spanish of the results
and implications of this and other meta-analyses of leadership
and gender). Recently, a meta-analysis of the studies—all
correlational—on gender differences and the transformational
and transactional styles has been published (Eagly et al.,
57
2003). The general results show that women, in comparison
to men, are more transformational and they are more
committed to behaviors involving contingent reward (a
component of transactional leadership). In contrast, male
leaders display behaviors that represent passive leadership
styles (direction by exception and laissez-faire) more
frequently than do females.
Many studies have been carried out on the congruity
between the typification of a gender role or activity and the
sex of the person who performs it, synthesized in the metaanalysis of Davison and Burke (2000). Their results
confirmed, in line with the role congruity theory of prejudice
(Eagly & Karau, 2002), that men and women were evaluated
less favorably when they were selected to perform tasks that
were incongruent with their gender. Given that leadership
roles are male sex-typed, women are evaluated less favorably
than men in these roles. Subsequent works that focused on
the analysis of this aspect of the theory (García-Retamero
& López-Zafra, 2006) have experimentally confirmed that
women are less likely than men to be promoted to a
leadership positions, especially in settings that are
incongruent with their gender role. Along the same line,
Ritter and Yoder (2004) investigated experimentally the
effect of the leader’s (male or female) dominance, the type
of task (male, female, or neutral), and the gender (same-sex
or mixed dyads) on the emergence of leadership. Their
results confirmed that, although women possess the agentic
quality of dominance that is consistent with the leader’s
role, incongruity between masculinized task demands and
gender stereotypes prevent women from emerging as leaders.
On the other hand, the findings of the correlational study
of Davis (2004) disprove Eagly and Karau’s assumption that
women leaders are evaluated as less effective than men in
masculine work settings.
Other works have explicitly evaluated the congruity
between context and leadership styles. For example, Gardiner
and Tiggeman (1999) investigated the adoption of taskoriented and relationship-oriented leadership styles in
organizations numerically dominated either by men or by
women. Their results showed that, in organizations
numerically dominated by men, the women adopted a malestereotypical leadership style and inhibit behaviors that are
typical of “interpersonally-oriented” styles because of the
negative consequences involved. Subsequently, van Engen,
van der Leeden, and Willemsen (2001) investigated the
influence of gender typification of jobs on leadership styles
adopted by men and women. However, their results showed
that there was no relation between the investigated leadership
styles (task/relationship) and whether the context was
masculine or feminine. In Spain, Cuadrado, Navas, and
Molero (2004) studied the congruity between the leader’s
sex, gender typification of the organization, and leadership
style. The results revealed that, according to their
subordinates, women displayed “feminine” leadership
behaviors to a greater extent in organizations “numerically
58
CUADRADO, MORALES, AND RECIO
dominated by women” and dedicated to “typically feminine”
activities than in the opposite case (male leaders, in
organizations “numerically dominated by men” and dedicated
to “typically masculine” activities). Recently, Dulewicz and
Higgs (2005) investigated leadership style in relation to the
leaders’ sector, gender, function, nationality, and personality
traits. The results revealed that the leadership styles were
independent of these variables.
Lastly, congruity between leadership style and the
leader’s sex has been investigated, but separately, to be
precise, either with the autocratic versus democratic styles
(see Luthar, 1996), or with task-oriented and relationshiporiented styles (see Rojhan & Willemsen, 1994). In the first
study, in contrast to the theory defended by Eagly and Karau
(2002), it was experimentally shown that female leaders
who adopted an autocratic style were evaluated more
positively than male leaders who used this style. In the
second study, the congruity between leadership style and
the leader’s sex did not produce effects in men’s and
women’s general evaluations of leadership. It was merely
found that male evaluators devalued men and women who
used styles that were incongruent with their gender (women
who adopted the task-oriented style and men who adopted
the relationship-oriented style) and only in measures of
efficacy (not in measures of leader’s likeability). For the
female evaluators, the evaluation did not vary as a function
of the leader’s sex and the style they adopted.
As can be seen, there are many works related to the goal
of this article, but most of them do not address the issue
directly. Some studies investigate the differences between
men and women in leadership styles; others focus on the
congruity between work setting—typically masculine or
feminine—and sex of the occupants (independently of the
leadership styles). In other works, the congruity between
gender typification of the context and leadership styles was
studied. In all these investigations, correlational or
experimental designs were used, and the results obtained
do not always point in the same direction.
Our goal is to verify experimentally one of the
assumptions of the role congruity theory of prejudice toward
female leaders (Eagly & Karau, 2002): Women who occupy
leadership roles are evaluated more unfavorably than men
when they adopt stereotypically masculine styles. We shall
also verify whether there are gender biases against women
compared to men in the leadership evaluations they receive.
For this purpose, and in contrast to previous
investigations, this study simultaneously includes the
leadership styles most frequently studied from the
psychosocial perspective, which have a stereotypically
masculine or feminine nature, according to the literature. In
other words, this study is not based on the differences
between men and women, but on the styles related to the
masculine or feminine dimension of gender stereotypes.
Specifically, we study experimentally the effects of congruity
between leadership styles (masculine—autocratic and task-
oriented—and feminine—democratic, relationship-oriented,
and individualized consideration) and the leader’s sex,
choosing a neutral context with regard to gender and
maintaining the context constant. From all the above
considerations, we formulate the following hypotheses:
Hypothesis 1: Female leaders will receive less favorable
evaluations than male leaders when they adopt
stereotypically masculine leadership styles (autocratic
and task-oriented).
Hypothesis 2: Male leaders will not receive less favorable
evaluations than female leaders when they adopt
stereotypically feminine leadership behaviors
(democratic, relationship-oriented, individualized
consideration).
Hypothesis 3: Female leaders will receive worse
evaluations from male evaluators than from female
evaluators.
Hypothesis 4: Male leaders will receive similar
evaluations from male and from female evaluators.
Method
Participants
The sample was made up of 136 psychology students
from the Universidad Nacional de Educación a Distancia
(National Open University of Spain), from the entire Spanish
territory, of whom 53% were women (mean age 27 years,
SD = 6.6) and the remaining 47% were men (mean age 29
years, SD = 6.8). The participants were assigned randomly
to each experimental condition (34 in each group),
confirming group homogeneity with regard to sex. The only
information provided about the investigation was that it was
a study of decision making. Participants were in an
introductory course of the 2nd year of Social Psychology
and were not familiar with the contents addressed in the
study. Task duration was approximately 20 minutes. The
participants received credit for their participation.
Measures, Experimental Design, and Procedure
Investigations on the effects of gender stereotypes on
evaluations of leader usually use narrations that describe
the leaders’ behaviors, varying the leader’s sex and keeping
the remaining characteristics constant (Rojahn & Willemsen,
1994). All such studies in which men’s and women’s
behavior is kept constant, only varying the leader’s sex, are
variants of Goldberg’s (1968) paradigm, the most frequently
used to verify whether women are at a disadvantage with
regard to men of the same characteristics. The most
frequently employed procedure has been to present a case
in which one half of the participants evaluate the leader
with a woman’s name and the other half evaluates the leader
with a man’s name (Eagly & Karau, 2002, p. 582). Our
WOMEN’S ACCESS TO MANAGERIAL POSITIONS
work falls within the framework of this paradigm and, in
addition to this variable, we also manipulated leadership
style.
We elaborated a narrative in which we manipulated the
leader’s sex (male vs. female) and leadership style (malestereotypical, or autocratic and task-oriented vs. femalestereotypical: democratic, relationship-oriented, and
individualized consideration). This narrative was especially
written for this investigation. In it, participants are urged to
imagine that they are part of an emergency service in a
public health hospital. We chose this context because it is
neutral regarding gender, in the sense that it is not typed as
either masculine or feminine. Gender typification of an
activity or context is based, on the one hand, on the
numerical representation of each sex in the various
occupations and, on the other, by the fact that either men
or women are considered more suitable to perform activities
that require stereotypically masculine or feminine
characteristics, respectively (Davison & Burke, 2000). Taking
these observations into account, gender stereotypes are not
easily activated in the setting of an emergency service of a
public hospital (as could occur in an automobile factory or
a beauty center). In fact, the context selected is not gendertyped in the meta-analyses of Eagly and Johnson (1990),
Davison and Burke, or in the study by Cuadrado et al. (2004)
carried out in Spain. On the other hand, this context is
occupied both by doctors—a stereotypically masculine
profession—and by nurses—a stereotypically feminine
profession. This statement is based empirically on the
percentages of men and women dedicated to these
professions. In medicine, the presence of women has
currently increased a great deal compared to the past, when
it was considered a stereotypically masculine activity. Despite
this, the percentage of female medical doctors is still lower
than that of males (it was 37.82% in 2000, and 42.10% in
2005). In contrast, nursing was and still is being performed,
for the most part, by women—around 94% of women from
2000 to 2005 (Instituto de la Mujer [Women’s Institute],
2006).
Participants were asked to evaluate a supervisor (male
or female, depending on the experimental condition) of the
service, who was occupying that positions for a trial period.
In the narrative, after reading the description of the
supervisor’s behavior, they were asked to collaborate in the
evaluation process by means of an anonymous questionnaire
in order to facilitate making decisions about the candidates.
Four different versions of the same narrative were
written, varying the leader’s sex (male vs. female) and their
leadership style (stereotypically masculine vs. stereotypically
feminine). The narrative, which, except for the experimental
manipulation of the variables, is similar in the four versions,
describes the supervisor’s behavior. The female leader’s
name was Lucía and the male leader’s name was Carlos.
Autocratic and democratic leadership styles were
operationalized in the versions from definitions of classic
59
(Lewin, 1939/1964; White & Lippitt, 1960) and
contemporary literature (Eagly & Johnson, 1990; Luthar,
1996). In order to operationalize the task- and relationshiporiented styles, we used the study of Yukl (1999) as our
starting point. Lastly, the operationalization of the
transformational style of individualized consideration was
based on the definition of this factor proposed by Bass and
Avolio (1990), presented in the introduction. Several
leadership experts reviewed the versions and their interrater
agreement about the clear depiction of the leadership styles
in the stories was confirmed. We reproduce below the part
of the cases presented which refers to the operationalization
of the variables in two of the experimental conditions (female
leader with stereotypically feminine style and male leader
with stereotypically masculine style):
“Lucía called for a first meeting in which she presented
all the information she had about the problem of user
satisfaction. At this meeting, she encouraged the team members
to express their opinions, to contribute their ideas, and to
suggest possible actions to solve the problem. She appreciated
and always took all the team members’ suggestions into
account. At this time, there were several internal arguments
among various members of team, which Lucía tried to prevent.
She offered encouragement and support when the situation
became particularly difficult. Lucía allowed each team member
to work with whomever they pleased, favoring the relations
among the entire team. Lucía treated all the team members
individually and made them feel that their contribution to the
improvement of the service quality was important.”
“Carlos called for a first meeting in which he presented
all the information he had about the problem of user
satisfaction. At this meeting, he informed the team about the
activities he had decided to carry out to solve the problem.
Carlos explained at the beginning the new tasks the team would
have to carry out and indicated how to perform them. He
planned in detail how to achieve each task, and the
responsibilities of each team member and the specific results
he expected. He explained how to organize and coordinate the
work activities to avoid delays, duplication of efforts, and
wasting resources. Carlos treated all the team members like
the components of a group and he made them feel that the
contribution of the group in general to the improvement of the
service quality was important.”
Therefore, the experimental manipulation of the leader’s
sex (male vs. female) and leadership style (stereotypically
masculine vs. stereotypically feminine) led to a 2 × 2 (Sex
× Style) factor design with 34 subjects in each experimental
condition. Neither leadership style nor leader’s sex were
measured as intrasubject variables, because the participants
could easily find out the goals of the investigation and, thus,
bias the results.
After reading the narrative, the participants had to
complete various measures:
60
CUADRADO, MORALES, AND RECIO
1. A list of 14 adjectives (7 positive and 7 negative),
selected from the study of Rudman and Kilianski
(2000). Participants rated the degree to which they
thought these adjectives were applicable to the
supervisor, using a 7-point Likert-type frequency
category scale, ranging from 1 (never) to 7 (always).
The positive adjectives were: competent, intelligent,
loyal, honest, clever, responsible, and optimistic. The
negative adjectives were: careless, forgetful,
discouraging, harmful for the team, bossy, snobbish,
and dishonest. These adjectives were used to
determine the direction of the evaluation and the
image the participants had formed of the leader.
2. Supervisor’s leadership capacity. A measure of 4 items
selected from the study of Luthar (1996) and of
Rojhan and Willemsen (1994), with a 7-point Likerttype scale, ranging from 1 (totally negative/totally
disagree) to 7 (totally positive/totally agree). The
items presented were: How would you evaluate X’s
general leadership capacity? X is a competent
supervisor. How would you evaluate X’s capacities
and aptitudes as a supervisor? How would you
evaluate X as a leader or supervisor?
3. Leadership effectiveness of the supervisor. A measure
of 5 items selected from the study of Luthar (1996)
and of Rojhan and Willemsen (1994), with a 7-point
Likert-type scale ranging from 1 (totally
negative/totally disagree) to 7 (totally positive/totally
agree).. The items presented were: How would you
evaluate X’s general achievement as a supervisor? X
is an effective manager. X knows how to manage
people effectively. X has sufficient capacity to be
considered a successful leader. X does not perform
his/her work as supervisor well enough.
Finally, we included seven manipulation-checks to verify
that the experimental manipulation had had some effect.
A pilot study was carried out to test the functioning of
all the aspects of the investigation, in particular, the narratives
and dependent variables. In this pilot study, 40 psychology
students from the Universidad de Almería (75% women and
25% men) participated, mean age 22.1 years (SD = 9.6). To
evaluate the different versions of the narrative, a 7-point
Likert scale, ranging from 1 (completely disagree) to 7
(completely agree) was used where participants rated the
credibility, concision, clarity, and truth of the information
presented, both in the general narrative and in the different
versions. In all the items, the median of the scores
corresponded to the value 5 (agree a lot) or 6 (agree very
much), so, in general, the participants considered that the
information presented was credible, concise, clear, and, true.
With regard to the psychometric quality of the measures of
the dependent variable, reliability was calculated with the
alpha coefficient, finding values of .86 for the items of
leadership capacity, .89 for the items of efficacy in leadership
performance, and .89 for the list of adjectives.
Data Analysis
After having confirmed that the pertinent assumptions
were met to perform parametric tests on the definite
sample, we analyzed the mean differences (Student’s t test)
between the leadership styles in the seven manipulationcheck items to determine whether the experimental
manipulation of the independent variable leadership style
had had effect. Likewise, we analyzed the reliability of
the measures used to estimate leadership with the alpha
coefficient. To verify the research hypotheses, we conducted
two ANOVAs: a 2 × 2 (Leader’s sex × Leadership style)
ANOVA for Hypotheses 1 and 2, and a 2 × 2 (Leader’s
sex × Evaluator’s sex) ANOVA for Hypotheses 3 and 4.
All the analyses were performed with the statistical package
SPSS 12.0.
Results
Manipulation Check
We compared the responses of the participants assigned
to the “stereotypically masculine” condition with those
assigned to the “stereotypically feminine” condition in the
seven manipulation-check items inserted in the questionnaire.
The analyses confirmed the validity of the experimental
manipulation (see Table 1). For example, in the first control
item, “X has a style that is basically oriented to the team
members,” the mean of the participants who read the
narrative about a leader with a stereotypically masculine
style was significantly lower than the mean of the
participants who read the narrative about a stereotypically
feminine style.
All the mean differences were statistically significant at
p < .01, except for those of the item “X’s style can be
considered stereotypically feminine,” which was statistically
significant at p < .05.
Reliability of the Dependent Variables
We calculated the alpha coefficient, finding a = .85 for
the adjective list, α = .90 for items of leadership capacity,
and α = .91 for items of leadership efficacy. The correlation
between the total score in leadership capacity and leadership
efficacy was .88, which provides evidence of the validity
of both scales, given the proximity of the terms.
Testing the Study Hypotheses
As displayed in Table 2, the ANOVA revealed a
significant main effect of the variable leadership style on
the three dependent variables, F(1, 126) = 70.34, F(1, 126)
= 45.15, and F(1, 126) = 29.10, for the adjective list,
leadership capacity, and leadership efficacy, respectively (p
61
WOMEN’S ACCESS TO MANAGERIAL POSITIONS
Table 1
Mean Differences between the Leadership Styles in the Manipulation-Check Items
Items
Stereotypically
masculine style
M
Stereotypically
feminine style
M
t(134)
1.88
5.21
–13.86**
4.38
3.38
3.59**
6.18
4.16
7.61**
5.85
1.9
13.47**
2.93
3.57
2.61*
1.78
5.76
–15.91**
1.88
5.72
–13.46**
1. X’s style is basically oriented to members of the team
(relationship-oriented; SF)
2. X’s style can be considered stereotypically male
3. X’s style is basically oriented to achieving the task goals
(task-oriented; SM)
4. X does not allow the team to participate in the decision
making (autocratic; SM)
5. X’s style can be considered stereotypically female
6. X promotes the team’s participation in the decision
making (democratic; SF)
7. X considers the team members individually
(individualized consideration; SF)
Note. SM = stereotypically masculine; SF = stereotypically feminine; X = leader’s name (Lucía or Carlos).
*p < .05. **p < .01.
Table 2
Means and Standard Deviations (in Brackets) and Effects of Leader’s Sex, Leadership Style, and their Interaction on the
Adjective List, Leadership Capacity, and Leadership Efficacy
Leader’s Sex
Dependent
variables
Adjective
list
Leadership
capacity
Leadership
efficacy
Sex × Style Interaction
Hypothesis 1
Hypothesis 2
Leadership Style
Male
M (SD)
Female
M (SD)
F(1, 126)
SM
M (SD)
SF
M (SD)
F(1, 126)
Male SM Female SM
M (SD)
M (SD)
Male SF
M (SD)
4.92 (.64)
4.90 (1.03)
0.02
4.32 (.73)
5.51 (.87)
4.76 (1.52)
4.73 (1.39)
0.02
4.00 (1.26)
4.87 (1.46)
4.74 (1.32)
0.34
4.21 (1.36)
Female SF
F(1, 126)
M (SD)
70.34**
4.30 (.64)
4.34 (.82)
5.55 (.84)
5.47 (.91)
0.21
5.48 (1.24)
45.15**
3.83 (1.39)
4.17 (1.13)
5.63 (1.11)
5.36 (1.36)
1.25
5.41 (1.15)
29.10**
4.18 (1.56)
4.21 (1.14)
5.52 (1.01)
5.33 (1.29)
0.22
Note: SM = stereotypically masculine; SF = stereotypically feminine.
**p < .01.
< .01 in all cases). However, as can be seen, no effect of
the variable leader’s sex or of the interaction of these
variables was found.
Thus, the results show that, independently of their sex,
when leaders adopt a stereotypically feminine leadership
style, they obtain significantly more favorable evaluations
on the adjective list, in the items of leadership capacity, and
in the items that measure leadership efficacy (see Figure 1).
Therefore, the results of the study reject the devaluation of
women when they adopt a masculine leadership style, as
this style received the least favorable evaluations in both
sexes. Hence, we found no empirical evidence to support
our first hypothesis.
However, the lack of differences in the evaluation
received by male and female leaders who adopt
stereotypically feminine styles supports our second
hypotheses (see Table 2). Nevertheless, this result should
be interpreted with caution, as it did not occur in
combination with the first hypothesis. Men did not receive
less favorable evaluations when they adopted stereotypically
feminine leadership styles (confirmation of Hypothesis 2),
but neither were the women devaluated when they managed
in stereotypically masculine ways (rejection of Hypothesis
1). Therefore, neither sex was evaluated unfavorably when
adopting gender incongruent styles (contrary to what could
be expected for female leaders).
62
CUADRADO, MORALES, AND RECIO
Figure 1. Effects of leadership style (SM = stereotypically
masculine; SF = stereotypically feminine) on the adjective list,
leadership capacity, and leadership efficacy.
With regard to Hypotheses 3 and 4, the ANOVA revealed
no main effects of either of the two independent variables
or of their interaction. The mean evaluation of the leaders
was practically identical in males and females on the
adjective list, F(1, 124) = .001, p = .98, in leadership
capacity, F(1, 124) = .013, p = .91, and in leadership
efficacy, F(1, 124) = .170, p = .68. Concerning the
evaluator’s sex, the evaluations made by men and women
were also similar in the adjective list, F(1, 124) = .02, p =
.88, in leadership capacity, F(1, 124) = .003, p = .88, and
in efficacy, F(1, 124) = .007, p = .94. The interaction of the
independent variables (leader’s sex and evaluator’s sex) did
not yield statistically significant results in the evaluation of
leadership in any of the three variables under consideration:
F(1, 124) = .001, p = .98; F(1, 124) = .166, p = .69, and
F(1, 124) = .117, p = .73, for the adjective list, leadership
capacity, and leadership efficacy, respectively.
Discussion
As mentioned, the main goal of our work was to verify
whether women are evaluated less favorably than men when
they manage in a stereotypically masculine way. The results
of our study reject this devaluation and show that this style
is actually evaluated less favorably in both sexes. Thus, the
main finding is that leaders, independently of their sex, receive
significantly better evaluations in the three measures included
(adjective list, leadership capacity, and performance efficacy)
when they adopt stereotypically feminine leadership styles
(versus stereotypically masculine styles). These results emerge
regardless of whether the evaluators are men or women.
These results have some important implications for the
literature on leadership and gender. Thus, it is not confirmed
that women in leadership positions who adopt stereotypically
masculine ways are less favorably evaluated than men (Eagly
& Karau, 2002). Like in other studies, the assumption on
which the role congruity theory of prejudice toward female
leaders is based is not confirmed (e.g., Davis, 2004;
Dulewicz & Higgs, 2005; van Engen et al., 2001; Rojhan
& Willemsen, 1994). What can be the explanation of these
results?
On the one hand, it is fairly likely that the context used
in this study (a hospital, which is not sex-typed) may have
influenced the results. These results suggest that, to the
extent to which activities and contexts cease to be
stereotyped as masculine or feminine, the evaluation of
female leaders will improve. On the other hand, the data on
which Eagly and Karau (2002) base this part of their theory
proceed from the meta-analysis of gender and evaluation of
leaders (Eagly et al., 1992). The changes that have occurred
since performing the studies that made up the meta-analysis
(the average year of publication was 1980) till the present
are, in our opinion, another key factor. As noted by Eagly
and Carli (2003b), “the increase in female leaders has been
accompanied by changes in theories and practices of
leadership” (p. 809). Thus, in psychosocial literature, the
existence of what is called the “female leadership advantage”
is debated: stereotypically feminine leadership styles are an
advantage for women because they are in consonance with
current organizational demands.
However, there are some studies (see, for example,
Cuadrado, 2004) that show that female leaders describe
themselves using instrumental traits and they assign more
importance to individualist values (both related to the
masculine dimension of gender stereotypes) to a greater
extent than do women who do not occupy leadership
positions. In other words, women who occupy leadership
positions continue to emulate masculine behavior,
traditionally associated with professional success in
leadership posts.
Fortunately, in the present study, as predicted by the role
congruity theory of prejudice toward female leaders (Eagly
& Karau, 2002), men are not evaluated less favorably than
women if they transgress the prescriptions associated with
their gender, namely, if they use stereotypically feminine
styles. Thus, Eagly and Karau acknowledge that over the
last years, there has been little change in the content of
gender stereotypes, but there has been a greater change,
tending toward androgyny, in the leadership profile
demanded by organizations (see Barberá et al., 2005). It is
fairly likely, therefore, that one of the consequences of this
evolution is that men become aware that they should let go
of the traditional style of direction and control and manage
in a more feminine or androgynous way because of the
positive consequences this may have in their performance
as leaders. This, in turn, might allow women to adopt
WOMEN’S ACCESS TO MANAGERIAL POSITIONS
leadership styles more congruent with their stereotype, as
well as to cease receiving unfavorable evaluations when
they add agentic (masculine) behaviors to their communal
(feminine) repertory. The results of this study indicate that
this is possible. Future investigations should study the results
obtained when using different participants and settings.
In short, our study reveals that female leaders do not
receive less favorable evaluations than males when they
adopt stereotypically masculine styles (i.e., autocratic), nor
is there a biased tendency in men to favor leaders of their
own sex, as Eagly et al. (Eagly, Makhijani, et al., 1992;
Eagly, Karau, et al., 1995) found in their meta-analyses. In
view of these results, one could wonder why women are
not sufficiently represented in leadership roles. In other
words, the scarcity of women in these positions is difficult
to understand if, as revealed by the results of this study and
other works (see Barberá & Ramos, 2004), the adoption of
stereotypically feminine styles is more highly valued than
masculine styles and more congruent with current
organizational demands.
Firstly, the existence of a persistent stereotype that
associates managerial activities with stereotypically male
qualities should be taken into account. Through diverse
studies in different countries, this phenomenon, called “think
manager-think male” in the literature, is global and is
especially sustained by men (see Schein, 2001). Therefore,
the traits, values, or behaviors linked to masculinity are
considered necessary to achieve success in managerial
positions. As a result, some women who occupy leadership
positions continue to emulate male behavior traditionally
associated with professional success in leadership positions
(see Cuadrado, 2004). However, when doing this, women
contravene desirable feminine behavior, namely, they
transgress the prescriptions associated with their gender and
are evaluated negatively, as was developed extensively in
the theory formulated by Eagly and Karau (2002). Because
of this, there is both a process of self-exclusion, known as
the “cement ceiling”—self-imposed by personal choices,
such as rejecting promotion (Chinchilla & León, 2004)—
as well as the existence of greater difficulty to achieve these
roles in comparison to men.
However, as aptly noted by Eagly (2003), the growing
presence of women in these positions, together with a higher
evaluation of stereotypically feminine qualities—which at
the same time are appropriate to the current needs of the
diverse settings (see the analysis of Barberá et al., 2005)—
are important factors that will favor female presence in
leadership positions. This study confirms these statements.
The changes—though slow—in the content of gender
stereotypes (see Barberá & Ramos, 2004) and the decrease
of incongruity between leadership roles and the feminine
role will allow organizations to be more receptive to female
leaders. Of course, this optimistic viewpoint should be
interpreted with realism. In fact, the statistics still show that
women have not yet advanced as much as would be fair
63
and desirable. There is still discriminatory treatment toward
female leaders or toward women who attempt to perform
these roles, but, from our viewpoint, some conditions must
be met for this to occur. For example, it is more likely to
occur in people who display strong gender stereotypes, and
when women try to gain access to positions of power and
prestige in jobs that are incongruent with their gender. We
hope these events will become more anecdotic and the results
obtained in this experiment will become a reflection of some
social advances on the way toward true equality.
Despite these favorable results, this investigation has
some limitations that are habitual in experimental studies
carried out with university students. Nevertheless, as
indicated by Eagly and Karau (2002, p. 587), the samples
of most experiments on leadership styles under Goldberg’s
paradigm are made up of university students. Another
limitation refers to external validity. However, according
to Dipboye and Flanagan (1979), limited external validity
is applicable both to experimental studies and to field
studies as, in both cases, the sample, situations, and
behaviors are selected. In this sense, our results would be
more generalizable if similar experiments were carried
out with different kinds of samples and in different
settings. Likewise, it would be interesting to confirm that
the results concerning the better evaluation of the feminine
leadership style are replicated using other experimental
designs.
While aware of the limitations, the main implication of
these results for the access of women to leadership positions
is that organizations should take into account the
contributions and the value of feminine styles, instead of
blocking women’s access to traditionally masculinized
settings. Likewise, they should consider the importance of
the feminine styles when training managerial skills and
styles. It is very likely that these aspects, together with the
growing access of women to leadership positions—which
will gradually modify the content of gender stereotypes—
will prevent the devaluation of female leaders and allow
egalitarian access of men and women to positions of
responsibility. Not only women, but organizations and society
in general can achieve important benefits.
References
Ayestarán, S. (2003). ¿Dirigen las mujeres igual que los hombres?
Revista de Psicología Social, 18, 315-319.
Barberá, E., Candela, C., López, M., Ramos, A., Sarrió, M.,
Albertos, P., Benítez, I.J., Díez, J.L. y Lacort, J.A. (2005).
Género y diversidad en un entorno de cambio. Valencia:
Universidad Politécnica de Valencia.
Barberá, E., & Ramos, A. (2004). Liderazgo y discriminación de
género. Revista de Psicología General y Aplicada, 57, 147160. Monographic issue: La psicología y el acceso de la mujer
a la función directiva. (Coords.: J.F. Morales & I. Cuadrado).
64
CUADRADO, MORALES, AND RECIO
Bass, B.M., & Avolio, B.J. (1990). Transformational leadership
development: Manual for the Multifactor Leadership
Questionnaire. Palo Alto, CA: Consulting Psychologists Press.
Cann, A., & Siegfried, W.D. (1990). Gender stereotypes and dimensions
of effective leader behaviour. Sex Roles, 23, 413-419.
Chinchilla, N., & León, C. (2004). La ambición femenina. Cómo
re-conciliar trabajo y familia. Madrid: Aguilar.
Cuadrado, I. (2001). Cuestiones teóricas y datos preliminares sobre
tres estilos de liderazgo. Revista de Psicología Social, 16, 131155.
Cuadrado, I. (2003). ¿Emplean hombres y mujeres diferentes estilos
de liderazgo? Análisis de la influencia de los estilos de liderazgo
en el acceso a los puestos de dirección. Revista de Psicología
Social, 18, 283-307.
Cuadrado, I. (2004). Valores y rasgos estereotípicos de género de
mujeres líderes. Psicothema, 16, 279-284.
Cuadrado, I., & Molero, F. (2002). Liderazgo transformacional y
género: autoevaluaciones de directivos y directivas españoles.
Revista de Psicología del Trabajo y de las Organizaciones, 18,
39-55.
Cuadrado, I., & Navas, M. (2000). La técnica del meta-análisis
aplicada al estudio del liderazgo y el género: resultados e
implicaciones. Revista de Psicología General y Aplicada, 53,
303-317.
Cuadrado, I., Navas, M., & Molero, F. (2004). El acceso de las mujeres
a puestos directivos: género, contexto organizacional y estilos de
liderazgo. Revista de Psicología General y Aplicada, 57, 181-192.
Monographic issue: La psicología y el acceso de la mujer a la
función directiva. (Coords.: J.F. Morales & I. Cuadrado).
Cuadrado, I., Navas, M., & Molero, F. (2006). Mujeres y liderazgo:
claves psicosociales del techo de cristal. Madrid: Sanz y Torres.
Davis, C.A. (2004). Gender stereotypes and 360-degree leadership
feedback ratings: An application of role congruity theory.
Doctoral dissertation. University of Michigan. University
Microfilms International.
Davison, H.K., & Burke, M.J. (2000). Sex discrimination in
simulated employment contexts: A meta-analytic investigation.
Journal of Vocational Behavior, 56, 225-248.
Dipboye, R.L., & Flanagan, M.F. (1979). Research settings in
industrial and organizational psychology: Are findings in the
field more generalizable than in the laboratory? American
Psychologist, 34, 141-150.
Dulewicz, V., & Higgs, M. (2005). Assessing leadership styles and
organizational context. Journal of Managerial Psychology, 20,
105-123.
Eagly, A.H. (2003). The rise of female leaders. Zeitschrift für
Sozialpsychologie, 34, 123-132.
Eagly, A.H., & Carli, L.L. (2003a). Finding gender advantage and
disadvantage: Systematic research integration is the solution.
Leadership Quarterly, 14, 851-859.
Eagly, A.H., & Carli, L.L. (2003b). The female leadership
advantage: An evaluation of the evidence. Leadership Quarterly,
14, 807-834.
Eagly, A.H., & Johannesen-Schmidt, M.C. (2001). The leadership
styles of women and men. Journal of Social Issues, 57, 781-797.
Eagly, A.H., Johannesen-Schmidt, M.C., & van Engen, M. (2003).
Transformational, transactional and laissez-faire leadership
styles: A meta-analysis comparing women and men.
Psychological Bulletin, 129, 569-591.
Eagly, A.H., & Johnson, B.T. (1990). Gender and leadership style:
A meta-analysis. Psychological Bulletin, 108, 233-256.
Eagly, A.H., & Karau, S.J. (2002). Role congruity theory of
prejudice toward female leaders. Psychological Review, 109,
573-598.
Eagly, A.H., Karau, S.J., & Makhijani, M.G. (1995). Gender and
the effectiveness of leaders: A meta-analysis. Psychological
Bulletin, 117, 125-145.
Eagly, A.H., Makhijani, M.G., & Klonsky, B.G. (1992). Gender
and the evaluation of leaders: A meta-analysis. Psychological
Bulletin, 111, 3-22.
García-Retamero, R., & López-Zafra, E. (2006). Congruencia de
rol de género y liderazgo: el papel de las atribuciones causales
sobre el éxito y el fracaso. Revista Latinoamericana de
Psicología, 38, 245-257.
Gardiner, M., & Tiggemann, M. (1999). Gender differences in
leadership style, job stress and mental health in male- and
female- dominated industries. Journal of Occupational and
Organizational Psychology, 72, 301-315.
Goldberg, P. (1968). Are women prejudiced against women?
Transaction, 5, 316-322.
Helgesen, S. (1990). The female advantage: Women’s’ ways of
leadership. New York: Doubleday Currency.
Instituto de la Mujer (2006). Las mujeres en cifras. Data obtained
on October 8 at http://www.mtas.es/mujer/mujeres/cifras/tablas/
W582.XLS
Lewin, K. (1939/1964). Experiments in social space. In D.
Cartwright (Ed.), Field theory in social science: Selected
theoretical papers by Kurt Lewin (pp. 71-83). New York: Harper
Torchbooks.
Luthar, H.K. (1996). Gender differences in evaluation of
performance and leadership ability: Autocratic vs. democratic
managers. Sex Roles, 35, 337-361.
Morales, J.F., & Cuadrado, I. (2004). Introducción: Teoría de
congruencia de rol del prejuicio hacia líderes femeninos. Revista
de Psicología General y Aplicada, 57, 135-146. Monographic
issue: La psicología y el acceso de la mujer a la función
directiva (Coords.: J.F. Morales & I. Cuadrado).
Moya, M. (2003). Sobre la existencia y el origen de las diferencias
en el liderazgo entre hombres y mujeres. Revista de Psicología
Social, 18, 321-325.
Munduate, L. (2003). Género y liderazgo: diferencias entre hombres
y mujeres en el acceso a los puestos directivos. Revista de
Psicología Social, 18, 309-314.
Ritter, B.A., & Yoder, J.D. (2004). Gender differences in leader
emergence persist even for dominant women: An updated
confirmation of role congruity theory. Psychology of Women
Quarterly, 28, 187-193.
Rojahn, K., & Willemsen, T.M. (1994). The evaluation of
effectiveness and likeability of gender-role congruent and
gender-role incongruent leaders. Sex Roles, 30, 109-119.
WOMEN’S ACCESS TO MANAGERIAL POSITIONS
Rudman, L.A., & Kilianski, S.E. (2000). Implicit and explicit
attitudes toward female authority. Personality and Social
Psychology Bulletin, 26, 1315-1328.
Schein, V.E. (2001). A global look at psychological barriers to
women’s progress in management. Journal of Social Issues,
57, 675-688.
van Engen, M. L., van der Leeden, R., & Willemsen, T.M. (2001).
Gender, context and leadership styles: A field study. Journal
of Occupational and Organizational Psychology, 74, 581598.
van Engen, M.L., & Willemsen, T.M. (2004). Sex and leadership
styles: A meta-analysis of research published in the 1990s.
Psychological Reports, 94, 3-18.
65
Vecchio, R.P. (2002). Leadership and gender advantage. Leadership
Quarterly, 13, 643-671.
Vecchio, R.P. (2003). In search of gender advantage. Leadership
Quarterly, 14, 835-850.
White, R.K., & Lippitt, R. (1960). Autocracy and democracy: An
experimental inquiry. New York: Harper.
Yukl, G. (1999). An evaluative essay on current conceptions of
effective leadership. European Journal of Work and
Organizational Psychology, 8, 33-48.
Received June 8, 2006
Revision received February 10, 2007
Accepted April 13, 2007